Re: new module 'isnan-nolibm'

[Bruno Haible]

Ben Pfaff wrote:
> I was under the impression that some platforms were missing isnan
> entirely, so that this was worth more than just "isnan() without
> libm".  But I can't confidently name a system that lacks it.

The package I announced two days ago shows that most systems have isnan()
as a function:

$ ./show-portability isnan
libc                freebsd-5.2.1
libc                freebsd-6.0
libc                irix-6.5
libc                macosx-10.3
libc                netbsd-3.0
libc                openbsd-3.8
libc                solaris-2.10
libc                solaris-2.4
libc                solaris-2.5.1
libc                solaris-2.6
libc                solaris-2.7
libc                solaris-2.8
libc                solaris-2.9
libcygwin           cygwin
libm                aix-4.3.2
libm                aix-5.1.0
libm                aix-5.1.0
libm                cygwin
libm                hpux-11.00
libm                hpux-11.11
libm                osf1-4.0d
libm                osf1-5.1a
libm                solaris-2.10
libm                solaris-2.4
libm                solaris-2.5.1
libm                solaris-2.6
libm                solaris-2.7
libm                solaris-2.8
libm                solaris-2.9
libroot             beos
MISSING in          mingw

and mingw has it as a macro:

    #define isnan(x) (sizeof (x) == sizeof (float) ? __isnanf (x)   \
                      : sizeof (x) == sizeof (double) ? __isnan (x) \
                      : __isnanl (x))

(confirming the point that Eric just made: that the 3 functions may have
completely different names than 'isnan', 'isnanf', 'isnanl').

Bruno